Several research studies suggest a secular decline in AD trends. However factors influencing negative secular trend in AD are unknown. Besides, racial differences in these secular trends also need to be established for African American minorities. To address these issues, we propose to use a large population-based study with cognitive tests of executive functioning, episodic memory, and overall global cognition and orientation in 10,801 participants, and clinical diagnosis of AD in 2,839 participants of African American and European American ethnicity to study age, race, and sex-specific decline in AD trends between 1993 and 2012, and reasons for a negative secular trend in AD. More specifically, the five objectives of this proposal are: (1) Demographic differences in AD trends aim: evaluate age-, sex-, and race-specific trends in AD using composite and individual tests of cognition, ADLS, and clinical diagnosis in prevalent and incident Alzheimer?s AD from 1993 to 2012; (2) Risk factors aim: understand the association of demographic characteristics, hypertension, stroke, diabetes, Body Mass Index (BMI), perceived stress, and depressive symptoms on AD trends using composite and individual tests of cognition, ADLS, and clinical diagnosis of prevalent and incident AD between 1993 and 2012. (3) Protective factors aim: examine the association of educational attainment, income, social engagement, neuroticism, extraversion, physical activity, and cognition enriching activities on AD trends using composite and individual tests of cognition, ADLS, and clinical diagnosis of prevalent and incident AD between 1993 and 2012. (4) Medications use aim: Examine how anti-hypertensive, statin, and diabetes medication, and multi-vitamin supplement use may be associated with incident AD, and how change in hypertension, statin, and diabetes medication use may be associated with AD trends using composite and individual tests of cognition, ADLS, and clinical diagnosis of AD between 1993 and 2012. In this proposal, we use a newly developed AD likelihood score (ADLS) to translate clinical diagnosis of AD in clinical sample that uses 19 neuropsychological tests to likelihood of AD in our population sample that uses 4 short-battery neuropsychological tests. Such an approach will drastically improve our power to detect potential reasons for secular decline in AD trends. In addition, our proposal examines demographic differences in AD trends, and putative risk factors and preventive factors that might influence AD trends. This proposal has the potential to make a large public health impact in potentially identifying factors that could be influencing AD trends leading to intervention studies that could be geared towards improving the identified factors in the general population.